The effect of transcutaneous nerve stimulation (TNS) on joint pain was studied in patients with rheumatoid arthritis. Three different forms of TNS were used in an attempt to evaluate the influence of placebo effects. The results were evaluated by means of loading tests and from the patients' own assessments of pain relief. High intensity TNS near the painful joint gave pain relief of varying degree in 95% of the patients, while TNS of low intensity and TNS applied at a remote site gave an improvement in 75% and 5% respectively. It seems quite clear that TNS is effective in reducing joint pain. There are a number of factors which intimate that this improvement is not merely a placebo effect.